Does Nutch crawl automatically when I add new pages to the website?
No, you have to recrawl or create the index from scratch.
Tt wont re-crawl automatically. You can do either of these:
make the parent page of the new url re-crawled so that the new url enters the crawldb and will be fetched in subsequent fetch round.
Add the new url directly to the crawldb via inject command.
You should do scheduled crawling to keep your data up-to-date.
Open source Java Job Schedulers
Related
May I know how to work in nutch if I am getting new urls daily to crawl new urls and how to store in crawldb.I am new to nutch please tell me the approach.
New URLs can be added to Nutch's CrawlDb at any time using the inject command. The newly added URLs are then fetched and processed in the next generate-fetch-update cycle.
My website changes very frequently, and I need a way to dynamically generate a new site map every day.
I tried to use sitemap.js but it requires me to give it specific urls for my site.
I am wondering if there's a way to have it crawl the site and generate a site map based on the urls it finds dynamically.
If not, is there any other server-side script that I can use to dynamically generate site maps?
Thanks
Your website does contain any backend? Or storage for data? Or is it only clean HTML and nothing more? If its any of the 1 or 2nd option, then you can just exctract it from this. Or you can
1. get your homepage, extract all urls, 2. omit those from other domains. 3. Repeat for links that you've stored. 4. Do not store duplicates.
I performed a successful crawl with url-1 in seed.txt and I could see the crawled data in MySQL database. Now when I tried to perform another fresh crawl by replacing url-1 with url-2 in seed.txt, the new crawl started with fetching step and the urls it was trying to fetch is of the old replaced url in seed.txt. I am not sure from where it picked up the old url.
I tried to check for hidden seed files, I didn't find any and there is only one folder urls/seed.txt in NUTCH_HOME/runtime/local where I run my crawl command. Please advise what might be the issue?
Your crawl database contains a list of URLs to crawl. Unless you delete the original crawl directory or create a new one as part of your new crawl, the original list of URLs will be used and extended with the new URL.
I crawl sites in nutch 1.3. now I want to delete a url from crawldb, how can I do this? how I read from crawldb? I want to see urls that exist in crawldb.
To read from crawlDb you can use the CrawlDBReader class(org.apache.nutch.crawl package). To delete/remove a url from the crawlDb you can use try using the CrawlDBMerger class(org.apache.nutch.crawl package) with the "-filter" option. But I suggest writing a Mapreduce to delete urls according to your needs.
Some site have url pattern as www.___.com/id=1 to www.___.com/id=1000. How can I crawl the site using nutch. Is there any wway to provide seed for fetching in range??
I think the easiest way would be to have a script to generate your initial list of urls.
no. you have inject them manually or using a script